Abstract:
PROBLEM TO BE SOLVED: To provide a high-function storage area controller which is connected in an n-way storage area configuration provided with a shared storage area. SOLUTION: A storage area controller in a network is equipped with an I/O execution component, a component for allocating the right of ownership of metadata concerned with data to an owner storage area controller node, a lock manager for controlling a lock of the metadata during I/O, and a storage area constitution element for transferring messages between storage area controller nodes to request the state of the metadata to allow a lock and also request the lock to be reset to send an unlock signal. Then I/O is carried out on data whose metadata are possessed by an arbitrary owner storage area controller node on condition that a metadata lock protocol controlled by the owner storage area controller is based upon, and consistent relation between an arbitrary copy of held data and the data is held upon occasion. COPYRIGHT: (C)2005,JPO&NCIPI
Abstract:
An apparatus for managing resource reclamation in data storage systems comprises: a volume deletion metadata recorder for recording metadata for one or more deleted volumes; a policy engine responsive to a predetermined policy rule to apply the policy rule to the metadata; and the policy engine initiating policy-controlled data storage space reclamation for the one or more deleted volumes. A volume reclaimer is responsive to the policy engine for reclaiming a data storage space from the one or more deleted volumes; and a resource allocator allocates the data storage space.
Abstract:
A method implemented in a computer infrastructure having computer executable code having programming instructions tangibly embodied on a computer readable storage medium. The programming instructions are operable to determine whether a target disk of a map contains data unavailable to a downstream disk from an upstream disk in a FlashCopy cascade and detect whether the downstream disk has a copy of the data. Additionally, the programming instructions are operable to copy the data from the target disk to the downstream disk, if the target disk of the map contains data unavailable to the downstream disk from the upstream disk and the downstream disk does not have the copy of the data. Furthermore, the programming instructions are operable to refrain from copying the data from the target disk to the downstream disk, if the target disk of the map does not contain data unavailable to the downstream disk from the upstream disk or the downstream disk does have the copy of the data. Moreover, the programming instructions are operable to remove the map from the FlashCopy cascade.
Abstract:
Storage control apparatus is operable in a network of storage controller nodes having an owner node operable to control locking of a region of data during I/O activity, a messaging component operable to request locks, grant locks, request release of locks and signal lock release. The apparatus comprises an I/O performing component to perform I/O on data owned by any owner node, subject to compliance with lock protocols controlled by the owner node. Any Flash Copy image of the region of data is maintained in a coherency relation with the region of data, and the I/O performing component is operable to cache a previous positive confirmation that the region of data has been left in a coherency relation with any said Flash Copy image, and to perform I/O activity on the basis of said previous positive confirmation.
Abstract:
A method, in a remote copy pair, for resynchronizing data held at a primary and secondary after the pair has undergone separation and reestablishment of a remote copy relationship includes the steps of sending a change record identifying data segments changed at the secondary from the secondary to the primary, merging the change record from the secondary with a change record identifying data segments changed at the primary into a merged change record; and copying all identified changed data segments indicated in the merged change record from the primary to the secondary. The method may be implemented in hardware or software or a combination thereof.
Abstract:
Various systems, processes, and products may be used to source clean cascaded volumes. In particular implementations, a system, process, and product for source cleaning cascaded volumes may include the ability to store a primary volume and a plurality of cascaded backup volumes. The system, process, and product may also include the ability to determine whether a write to one of the backup volumes is due to a background copy being made of another volume in the cascade and adjust a background copy indicator in a volume map for the backup volume if the write is due to a background copy and to determine whether a write to one of the backup volumes is due to a write to another volume in the cascade and adjust a write copy indicator in a volume map for the backup volume if the write is due to a write to another cascaded volume.
Abstract:
Recovering a data backup comprising a data image and a point-in-time copy of the data image. Provided is a virtual data image disk for storing a copy of the data image and a virtual snapshot disk for storing data of the point-in-time copy. The data image is copied 402 from a backup to the virtual data image disk and data from the point-in-time copy of the data image is copied 402 to the virtual snapshot disk. It is then determined 407 if the data of the point-in-time copy exists in the virtual data image disk and if it does not then copying 410 data from the point-in-time copy to the virtual snapshot disk and so avoiding data replication they by reducing the amount of storage space used. If it is determined that the data already exists in the virtual data image disk then the virtual snapshot disk is mapped 412 to that data of the virtual image disk.
Abstract:
Ein Verfahren und ein System zum Speicher-Checkpointing in einem System gespiegelter virtueller Maschinen werden bereitgestellt. Das Verfahren enthält eine Speicher-Steuereinheit, die eine modifizierende Operation am Speicher von einer virtuellen Maschine empfängt und die modifizierende Operation nichtlöschend ausführt, indem die Daten der modifizierenden Operation in einem Checkpointing-Bereich des Speichers gesichert werden und der gespeicherte vorherige Zustand der Daten bewahrt wird. Das Verfahren beinhaltet außerdem Empfangen einer Checkpoint-Benachrichtigung und Festschreiben von Daten der modifizierenden Operation von dem Checkpointing-Bereich und Freigeben des gespeicherten vorherigen Zustands der Daten. Wenn eine Failover-Benachrichtigung empfangen wird, setzt die Speicher-Steuereinheit den physischen Speicher zurück, damit er mit einem Checkpoint-Zustand übereinstimmt.
Abstract:
A data storage system comprises a controller, a first lower performance storage medium and a second higher performance storage medium. The controller is connected to the storage mediums and is arranged to control IO access to the storage mediums. The controller is further arranged to store an image on the first storage medium, initiate a copy function from the first storage medium to the second storage medium, direct all IO access for the image to the second storage medium, periodically age data from the second storage medium to the first storage medium, create a new empty bitmap for each period, and in response to an IO access for data in the image, update the latest bitmap to indicate that the data has been accessed and update the previous bitmaps to indicate that the data has not been accessed.